Invest with AI
00:00:36 · S3
[00:00:36] S3: This conversation is provided for informational and educational purposes only. The views expressed are those of the individual speakers as of the recording date, may differ from the views of their organizations, and may change without notice. Nothing discussed constitutes investment, legal, accounting, or tax advice. An offer or solicitation, or a recommendation or endorsement of any security, investment strategy, or financial product. Access to this conversation does not, by itself, create an advisory, fiduciary, or client relationship. Any company's, securities, or strategies discussed are presented solely for illustrative purposes. The speakers or their affiliates may hold positions in, provide services to, or otherwise have financial interests in the companies or securities mentioned, and those interests may change without notice. Any forecasts, targets, forward looking statements, model outputs, or hypothetical results are based on assumptions and information available as of the recording date. They involve inherent risks and uncertainties, may not reflect actual trading or investment results, and should not be relied upon as guarantees of future outcomes. Information obtained from third party sources is believed to be reliable, but its accuracy, completeness, and timeliness are not guaranteed. All investing involves risk, including volatility and the possible loss of principal. Past performance is not indicative of future results, and no investment strategy can assure a profit or protect against loss. Listeners should conduct their own due diligence and consult their own qualified financial, legal, accounting, and tax advisors before making any investment decision. Artificial intelligence systems and other analytical models may produce incomplete, inaccurate, or inconsistent results. Any model generated analysis should be independently evaluated and should not be used as the sole basis for an investment decision.
00:02:03 · Brett Caughran
[00:02:03] S2: Welcome to the next episode of Invest with AI, where we explore the intersection of fundamental investing and artificial intelligence. We have a great product market fit with the next guest on our podcast today, Doug Clinton of Intelligent Alpha, who has been investing with AI. So excited to have you on the pod, Doug, today and explore everything you've learned that's worked well, that hasn't worked well on your journey of investing with AI. Maybe to start, could you tell us a little bit more about Intelligent Alpha and what you're doing there?
00:02:37 · Doug Clinton
[00:02:37] S0: Absolutely, Brett. Good to be with you. Good to be with UK. And Intelligent Alpha is a company that I started, two years ago now officially. And it was born out of an experiment that started three years ago where once kinda hit the market late twenty twenty two, I was curious with a simple question, which is can ChatGPT beat the S and P 500, which a lot of human managers can't do? And so we started running some tests to see if that was possible, and the short answer was things looked very promising very early on. Maybe it's a little beginner's luck, but I was excited enough about what I was kind of seeing from very early LLMs and betting on the trajectory of these LLMs continue to get smarter and smarter over time that we wanted to launch an investment business that was really built around the idea of harnessing the intelligence from frontier models, allowing them to do investment analysis, and then ultimately, portfolio management. The one of the small piece I'd add about my background and kinda why we even played in this space in the first place is almost a decade ago, I started a venture firm called Deepwater. I'm still a partner there. I I run kind of both sides of the coin being a human allocator and also relying on AI to allocate at Intelligent Alpha. But we invest in early stage technology startups. We invest now in late stage tech startups and public equity at Deepwater. And so it was kind of this fun, natural testing ground where we've been in this world, we've been investing in AI companies, and Intelligent Alpha was the next logical progression for us from that business at Deepwater.
00:04:11 · Brett Caughran
[00:04:11] S2: Super, super interesting, and I have many, many questions about that. One of the first is how you thought about just the fundamentals of large language models, operate on lexical intensity, right? LLMs know that Paris is the capital of France because it's seen it 14,000,000 times in the training corpus. Know, when I've sort of in the past gone and asked ChatGPT for a portfolio of 10 stocks, it seems to operate on that same sense of lexical intensity that the stocks like Nvidia that are talked about a lot on the on the open web get cited as individual ideas in the portfolio. That could be a signal or that could not be often sort of over the cycle of markets is probably a negative signal. And retail driven markets, probably positive signal. How have you thought about that concept of lexical intensity prevalence in the training corpus and building a market beating portfolio?
00:05:05 · Doug Clinton
[00:05:05] S0: Yeah, I'd say very early on, one of the things we sort of realized was in '23, the paradigm for using LLMs effectively to invest was prompt engineering. I mean, that was all the rage then and really understanding what what is the context that you're providing to the models in terms of the task you want it to output. And so like you just said, Brett, if you just say, give me a stock of 10 portfolios, you're probably gonna get, I would say, eight of the 10 largest companies in the S and P 500 would be my my bet, which is fine. That might be an okay portfolio. But I think when we think about really using AI to make intelligent investments, that's not what we want. And so giving it a different framework where, you know, maybe you narrow the field to small caps, and maybe you define that within certain parameters for market cap. Maybe you pull in some external data from sources where maybe it's fundamental data, maybe it's your own research about a handful of companies. I think that's where it starts to get much more interesting using these language models and seeing how they sort of interpret what is all the data that's out there that humans are are looking at. Right? And and language models can process way more qualitative data and quantitative data than any human analyst can. So trying to figure out how can you get that right data and the right selection set into the models is kind of step number one. And then I think as I'm sure you guys have seen too, right, the frontier has evolved very quickly where it was all about prompt engineering a couple years ago. I think now it's about agentic workflows, you know, orchestration. And I think even beyond where we're at right now, the thing that we're getting really into is how do you sort of manage organizational context? You know, instead of just narrow task context, you know, you're trying to pick a large cap portfolio. That's one specific vertical that you might be working on. How do you look at the entire corpus of every portfolio you've ever generated? Or maybe if you have multiple team members generating different portfolios with their agents, how do you sort of understand the entire broad, knowledge base that's in your company and make it useful to the models when you're doing investment work.
00:07:15 · Khe Hy
[00:07:15] S1: Doug, on that point, did you mention kind of the evolution from prompt and it seems like we went from prompt engineering to context engineering to agentic workflows. And now with kind of the Fable like models kind of intent engineering or, you know, output based prompting, I'm curious how that last bit... I mean, we're still early in the Fable five days, but, has that last bit changed your approach?